Adiposity and insulin resistance are fundamental features in the pathogenesis of type 2 diabetes mellitus (DM). Emerging data suggest that sex-steroid hormones and adipocyte-derived hormones and cytokines ("adipokines") may be associated with type 2 DM risk and that some of these novel markers may exhibit a sexual dimorphism with respect to this risk. In two large prospective cohorts of men and women with archived blood specimens, we propose to investigate the roles of endogenous steroid hormones and adipokines in the development of type 2 DM. A total of 1,200 incident eases of type 2 DM will be identified and confirmed among 12,304 women in the Women's Health Study (WHS) and 11,130 men in the Physicians' Health Study II (PHS II). We will assay five steroid hormones (i.e., total and bioavailable testosterone, estradiol-17beta [E2], dehydroepiandrosterone sulfate [DHEAS], and sex hormone binding globulin [SHBG]), and three adipokines (tumor necrosis factor-alpha [TNF], adiponectin, and resistin). These biochemical markers can be reliably measured using frozen plasma samples. Applying state-of-the-art genotyping technology and statistical methods, we will define functional variants and determine the structure of haplotypes in nine relevant candidate genes and evaluate their roles as predictors for risk of type 2 DM. These genes will include hormone-metabolizing genes (aromatase [CYP19], androgen receptor, estrogen receptor alpha [ESR1], and SHBG), as well as those related to adiposity and insulin resistance (i.e., peroxisome proliferator-activated receptor-gamma [PPARgamma] and genes that respond to PPAR3' signaling, including TNFalpha, adiponectin, resistin, and the adipocyte-fatty acid binding protein [aP2]). Elucidation of interrelationships between these novel biochemical and genetic markers and the development of type 2 DM may help identify high risk individuals and suggest new treatment and/or prevention strategies, some of which may be sex-specific. Several unique features of these established cohort studies, including their identical prospective design and data collection procedures, high follow-up rates, availability of stored blood specimens, and cost efficiency, make these populations exceptional resources for the etiologic investigation of type 2 DM in women and men.